While it is well known that the choice of germplasm used in industrial plantations or restoration plantings of forest trees will have an impact on plantation productivity and profitability, there has been little appreciation of the flow-on consequences that the choice of tree germplasm has on the broader community of organisms that develops within the plantation.    

For part of his time prior to leaving the University of Tasmania,  Dr Robert Barbour (see separate article in Biobuzz 8) was undertaking research for an ARC Discovery grant which was looking at the flow-on (often termed ‘cascading’ or ‘extended genetic’) effects of genetic variation within the Tasmanian blue gum (Eucalyptus globulus) on dependent biodiversity.  This research was focused on testing whether genetic-based differences between the races of E. globulus affected the composition and richness of the communities of arthropods and fungi that developed, not only on the living trees themselves but in the associated litter and decaying logs derived from the trees. His study commenced in 2004 and, depending on the study, used 15 or 17 year old trees from the Gunns Ltd. base population progeny trial of E. globulus  at West Ridgley.  Trees from eight races were studied (10 families per race and two-trees per family; a total of 160 trees). The trees were growing in a common environment field trial and had been effectively randomized in space, so that any differences in the biodiversity observed between races could be attributed to genetics. We have long known that individual species (usually studied in the context of plantation pests) of fungi, arthropods and marsupials respond to genetic variation both between and within the races of E. globulus, but Bob's research was the first community-level study of the effects of intra-specific genetic variation within E. globulus. To our knowledge it is one of the most detailed community-level studies of any single forest tree species. The results from this study have now been accepted for publication in a series of four papers.

In the first paper (Barbour et al. 2009a), a symptoms-based approach was used to assess the abundance of 8 insect and 8 fungal taxa in the canopy of felled trees.  The variation in canopy community was compared with variation in foliar phytochemistry, foliar morphology and the near-infrared reflectance (NIR) of the leaves. Races of E. globulus were shown to support significantly different canopy communities, mainly due to differences between the Australian mainland races and those from the Bass Strait islands, with the Tasmanian races tending to be intermediate.  Five of the eight fungal and insect taxa assessed showed significant race effects, including the well known insect pests Gonipterus scutellatus (more damage on mainland races) and Paropsisterna agricola (less damage on mainland races). The genetic-based divergence between races in their foliar canopy community was positively correlated with differences in their foliar NIR spectra and leaf morphology.  In other words, the more two races differ genetically in their leaf phytochemistry and morphology the more the fungal and insect community they support in their canopy is likely to differ. Four morphological and secondary chemical traits were correlated with the racial divergence in the canopy community, the two most notable being the dry mass per unit area of the leaf and the concentration of  condensed tannins in the leaves.  

The second paper (Barbour et al. 2009b) reported a study of the genetic differences between the races of E. globulus in the characteristics of the bark on the tree trunk, as well as the macroarthropod community associated with the loose bark on the trunk.   Substantial genetically-based differences were found between races in the quantity and type of decorticating bark. Significant variation existed among trees of different races in the composition of the community of organisms associated with this bark. There was a two-fold difference in species richness (7–14 species) and abundance (22–55 individuals) among races. This community variation was tightly linked with genetically based variation in bark, with 60% of variation in community composition driven by bark characteristics. These community-level effects of tree genetics are expected to extend to higher trophic levels because of the extensive use of tree trunks as foraging zones by birds and marsupials.

The third paper (Barbour et al. in press, a) reports the existence of tree genetic effects on a leaf litter invertebrate community and soil characteristics.  It focused on just two of the E. globulus races previously studied that were known to be genetically divergent,  one from mainland Australia (Strzelecki Ranges) and the other from Tasmania (Southern Tasmania).   The apical canopies of the trees that had been felled previously (see above) were placed next to the stump from which they originated in the plantation; the leaf litter habitat which developed beneath the decaying canopies of known genotype was studied.  Pitfall trap sampling for invertebrates and linseed germination bioassays of soil were conducted within this habitat. Two key findings emerged. Firstly, assessment of 27 invertebrate orders (57,924 individuals) revealed significant race-level variation in leaf litter biodiversity (i.e. in community richness, abundance, composition and diversity). Secondly, considerable race-level differences in soil characteristics were evident based on linseed germination and growth responses. While the previous two papers demonstrated the consequences of genetic variation within forest trees for organisms that interact directly (i.e. proximally) with the living tree, these findings highlight the distal impacts that genetic variation within a tree species may have on biotic communities as well as soil properties.

The fourth paper (Barbour et al. in press, b) extended this study of distal effects to the macro-fungal community that developed on the decaying logs from the felled E. globulus trees. Logs of uniform size and shape from trees of known genotype were placed as designed grids within a native E. globulus forest.  After three years of natural colonisation, the presence of 62 macrofungal taxa were recorded from eight microhabitats on each log. The key factor found to drive macrofungal distribution and biodiversity on structurally uniform course woody debris was log-microhabitat, explaining 42% of the total variation in taxon-richness.  Differences between log-microhabitats appeared to be due to variation in aspect, substrate (bark vs wood) and area/time of exposure to colonisation. Despite genetic differences in wood and bark properties existing between the races of E. globulus studied, there was no significant effect of tree genetics on macrofungal community richness or composition.  While the previous studies have demonstrated that the genetics of a foundation tree species such as E. globulus can influence dependent communities, this was not found to be the case for the early log decay community and such affects are likely to be guild specific.


These papers have shown that in tree species such as E. globulus where large genetically-based differences exist in phenotypic traits, the choice of genetic material for planting, whether it be for restoration plantings or industrial forest estates, may affect the successional trajectory of the communities that ensue. For plantation forestry, this choice may influence pest-management regimes; for environmental plantings, it may influence biodiversity and environmental values. Understanding such flow-on effects is likely to become increasingly important in predicting the adaptive responses of forest trees - and their associated biota - to climate change.

References:

Barbour RC, Baker SC, O'Reilly-Wapstra JM, Harvest TMA, Potts BM (2009a) A footprint of tree-genetics on the biota of the forest floor. Oikos 118 (12), pp. 1917-1923.

Barbour RC, Storer MJ and Potts BM (2009b) Relative importance of tree genetics and microhabitat on macrofungal biodiversity on coarse woody debris.  Oecologia 160 (2), pp. 335-342.